> because you aren't doing huge numbers of repeated operations (as in ML)
In ML precision doesn't count so much. It's often possible that we train on 32bits and deploy the model on 16bits. In fact stochasticity is useful and we add it in by dropout or other things. You can drop any connection (or millions of them) in a neural net and get almost the same result. There have been papers that showed that reducing the number of bits used to represent weights might even improve the network because it has a regularising effect. The real brain is also stochastic.
In ML precision doesn't count so much. It's often possible that we train on 32bits and deploy the model on 16bits. In fact stochasticity is useful and we add it in by dropout or other things. You can drop any connection (or millions of them) in a neural net and get almost the same result. There have been papers that showed that reducing the number of bits used to represent weights might even improve the network because it has a regularising effect. The real brain is also stochastic.